U.S. patent application number 12/225915 was filed with the patent office on 2010-08-19 for fuel cell stack components and materials.
This patent application is currently assigned to BLOOM ENERGY CORPORATION. Invention is credited to Tad Armstrong, Emad El Batawi, Deepak Bose, Stephen Couse, Matthias Gottmann, Vlad Kalika, Patrick Munoz, Martin Perry.
Application Number | 20100209802 12/225915 |
Document ID | / |
Family ID | 39926222 |
Filed Date | 2010-08-19 |
United States Patent
Application |
20100209802 |
Kind Code |
A1 |
Armstrong; Tad ; et
al. |
August 19, 2010 |
FUEL CELL STACK COMPONENTS AND MATERIALS
Abstract
A plurality of fuel cell stack components, such as
interconnects, seals and bypass conductors are provided. The
components may be used in solid oxide fuel cell stacks or other
types of fuel cell stacks.
Inventors: |
Armstrong; Tad; (Burlingame,
CA) ; Batawi; Emad El; (Sunnyvale, CA) ; Bose;
Deepak; (Fremont, CA) ; Couse; Stephen;
(Sunnyvale, CA) ; Gottmann; Matthias; (Sunnyvale,
CA) ; Kalika; Vlad; (San Jose, CA) ; Munoz;
Patrick; (Milpitas, CA) ; Perry; Martin;
(Sunnyvale, CA) |
Correspondence
Address: |
FOLEY AND LARDNER LLP;SUITE 500
3000 K STREET NW
WASHINGTON
DC
20007
US
|
Assignee: |
BLOOM ENERGY CORPORATION
|
Family ID: |
39926222 |
Appl. No.: |
12/225915 |
Filed: |
April 2, 2007 |
PCT Filed: |
April 2, 2007 |
PCT NO: |
PCT/US07/08224 |
371 Date: |
July 8, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60788043 |
Apr 3, 2006 |
|
|
|
Current U.S.
Class: |
429/469 ;
429/467 |
Current CPC
Class: |
H01M 8/026 20130101;
H01M 8/2425 20130101; H01M 8/0265 20130101; H01M 8/2432 20160201;
Y02E 60/50 20130101; H01M 4/8605 20130101; H01M 8/0282 20130101;
H01M 2008/1293 20130101; H01M 8/2483 20160201 |
Class at
Publication: |
429/469 ;
429/467 |
International
Class: |
H01M 8/24 20060101
H01M008/24 |
Claims
1. A fuel cell stack, comprising: at least one fuel cell; at least
one interconnect; and at least one compliant layer located between
the fuel cell and the interconnect, wherein the at least one
compliant layer functions as an adhesive between the at least one
fuel cell and the at least one interconnect.
2. The stack of claim 1, wherein the at least one compliant layer
comprises at least one of a seal and a contact layer.
3. The stack claims 2, wherein the at least one compliant layer is
applied as a tape, a paste, a foam, a felt or a mesh.
4. The stack of claims 3, wherein at least one of a binder or a
solvent in the compliant layer acts as an adhesive.
5. The stack of claim 4, wherein the at least one of binder or
solvent is burned out from the stack prior to operation of the
stack to generate electricity such that the compliant layer
functions as a temporary adhesive during stack assembly.
6. The stack of claim 4, wherein the compliant layer comprises a
heat activatable adhesive.
7. The stack of claim 4, wherein an adhesive material is applied to
at least one surface of the at least one compliant layer.
8. The stack of claim 4, wherein an adhesive material itself acts
as the compliant layer.
9. An interconnect for a fuel cell, comprising a flow separator
plate, a cathode contact media and an anode contact media, wherein
at least one of the contact medias is corrugated.
10. A fuel cell stack, comprising: a plurality of fuel cells
separated by a plurality conductive interconnects; and at least one
material which provides a conductive path between adjacent
interconnects when a cell between the adjacent interconnect
fails.
11. The stack of claim 10, wherein the at least one material which
provides a conductive path comprises at least one of: (a) a
metallic conductor that does not oxidize which is located in a cell
anode electrode; (b) a conductive oxide, nitride or carbide which
is located in a cell anode electrode; (c) a cell anode electrode
material which forms a conductive oxide upon oxidation of a metal
phase of a cermet anode electrode material; (d) a non-conductive
oxide that becomes conductive by application of a potential; (e) a
material which forms a conductive path by at least one of reduction
or dielectric break down; or (f) a resistor provided in parallel to
a fuel cell.
12. The stack of claim 10, wherein the material comprises a bypass
conductor located peripherally from the cell and which becomes
electrically conductive in response to a failure of the fuel
cell.
13. The stack of claim 12, wherein the bypass conductor comprises
an oxygen ionically conductive but electrically insulating material
which is located in electrical contact with the adjacent
interconnects.
14. The stack of claim 13, wherein the oxygen ionically conductive
but electrically insulating material becomes electrically
conductive by reduction in response to a failure of a fuel cell
located between the adjacent interconnects.
15. The stack of claim 14, wherein the material comprises a doped
ceria or a doped zirconia which is in electrical contact with a
cathode side interconnect.
16. The stack of claim 15, wherein the bypass conductor further
comprises an electrically conductive ceramic material in physical
contact with the doped ceria or doped zirconia and also in
electrical contact with an anode side interconnect.
17. The stack of claim 16, wherein the electrically conductive
ceramic material comprises LSM and the doped zirconia comprises
YSZ.
18. An interconnect for a fuel cell, comprising a flow separator
plate and an anode contact media, wherein the anode contact media
comprises a foam containing surface features which are configured
to direct a fuel gas flow.
19. The interconnect of claim 18, wherein the surface features
comprise ribs or grooves which are concave facing a fuel inlet and
are convex facing a fuel outlet.
20. An interconnect for a fuel cell, comprising a flow separator
plate and a cathode contact media, wherein the cathode contact
media comprises an oxidation resistant nickel alloy foam.
21. The interconnect of claim 20, wherein the nickel alloy foam
comprises 10 to 30 weight percent Cr, 10 to 20 weigh percent W, 0
to 5 weight percent Mo, 0 to 5 weight percent Co, at least 45
weight percent Ni, 0 to 5 weight percent Fe and 0 to 1 weight
percent of one or more of Mn, Si, Al, C, La and B.
22. The interconnect of claim 20, wherein the nickel alloy foam
comprises a Haynes 320 alloy foam.
23. A ceramic interconnect for a fuel cell, comprising: a plurality
of laminated ceramic layers; a hydrocarbon fuel reformation
catalyst material; and air and fuel flow fields on opposite sides
of the interconnect comprise offset holes in adjacent layers, such
that an offset hole-pattern results in an over-under air and fuel
flow paths.
24. The interconnect of claim 23, further comprising a plurality of
vias filled with an electrically conductive material.
25. The interconnect of claim 24, wherein: the interconnect
comprises at least five laminated ceramic layers; the first and the
second layers form a cathode side flow field in which openings in
the first layer are offset from the openings in the second layer,
such that a channel is formed under the first layer; the fourth and
the fifth layers form an anode side flow field in which openings in
the fourth layer are offset from the openings in the fifth layer,
such that a channel is formed under the fifth layer; and the third
layer is located between the second and the fourth layers and
comprises a gas separator layer.
26. An interconnect for a fuel cell, comprising at least one
hydrocarbon fuel reformation catalyst coated channel below a
surface of the interconnect which allows a fuel or air gas stream
to flow beneath the surface of the interconnect.
27. A method of operating a fuel cell stack comprising a plurality
of fuel cells separated by a plurality of interconnects, the method
comprising providing a hydrocarbon fuel stream beneath a catalyst
coated surface of the interconnect.
28. The method of claim 27, wherein the at least one of a fuel or
air gas streams flows above a first portion of the interconnect and
beneath a second portion of the interconnect.
29. The method of claim 27, further comprising reforming the
hydrocarbon fuel at the catalyst coated surface of the
interconnect.
30. A fuel cell stack seal comprising a ceramic weak boundary layer
located between two glass layers.
31. The seal of claim 30, wherein the weak boundary layer comprises
ceramic particles.
32. The seal of claim 31, wherein the weak boundary layer is
configured to preferentially fracture as the stack heats up and
cools compared to fuel cells of the stack.
33. The seal of claim 30, wherein the seal is located between a
solid oxide fuel cell and an interconnect in a solid oxide fuel
cell stack.
Description
[0001] This application claims benefit of priority of U.S.
provisional application No. 60/788,043 filed on Apr. 3, 2006, which
is incorporated herein by reference in its entirety.
BACKGROUND OF THE INVENTION
[0002] The present invention is generally directed to fuel cells
and more specifically to high temperature fuel cell systems and
their operation.
[0003] Fuel cells are electrochemical devices which can convert
energy stored in fuels to electrical energy with high efficiencies.
High temperature fuel cells include solid oxide and molten
carbonate fuel cells. These fuel cells may operate using hydrogen
and/or hydrocarbon fuels. There are classes of fuel cells, such as
the solid oxide regenerative fuel cells, that also allow reversed
operation, such that oxidized fuel can be reduced back to
unoxidized fuel using electrical energy as an input.
SUMMARY
[0004] Embodiments of fuel cell stack components and materials are
described herein. It should be noted that each embodiment can be
used independently of the other embodiments or together with any
one, two, three, four, five or six other embodiments described
below. Thus a fuel cell stack may include any combination of one to
seven of the below described embodiments. Furthermore, while the
embodiments are preferably used with solid oxide fuel cells (SOFC),
other high and/or low temperature fuel cell types, such as molten
carbonate, PEM, phosphoric acid, etc, may also be used if
appropriate.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] FIG. 1 illustrates a schematic side cross sectional view of
a portion of the fuel cell stack with a bypass conductor according
to the second embodiment.
[0006] FIGS. 2 and 3 are three dimensional and top views,
respectively, of fuel cell stack components according to the third
embodiment.
[0007] FIGS. 4-5 are three dimensional views of interconnect
components according to the fourth embodiment.
[0008] FIG. 6 is a computer simulator of the interconnect component
shown in FIG. 5.
[0009] FIG. 7 is a top view of an interconnect according to the
sixth embodiment.
[0010] FIGS. 8-9 are three dimensional views of an interconnect
according to the sixth embodiment.
[0011] FIGS. 10-11 are side cross sectional views of an
interconnect according to the sixth embodiment.
[0012] FIG. 12 is a side cross sectional view of a portion of a
fuel cell stack according to the seventh embodiment.
DETAILED DESCRIPTION OF THE EMBODIMENTS
I. Innovative Materials for Assembly of High Temperature Fuel
Cells
[0013] The first embodiment of the invention describes how the
compliant layers within a high temperature fuel cell stack can
serve multiple purposes.
[0014] Planar high temperature fuel cell stacks are usually
assembled using compliant layers for contact and sealing. These
compliant layers can serve multiple purposes during assembly.
[0015] Planar fuel cell stacks are usually assembled from
alternating layers of cells (electrolyte with electrodes (i.e.,
fuel and air electrodes on either side of the electrolyte) and
optionally a frame) and interconnects (single or multi-part). The
interconnect provides gas manifolding and electrical conductivity
between adjacent cells. In other words, the interconnect can also
serve as a gas separator plate. In a solid oxide fuel cell, the
electrolyte comprises a solid oxide (i.e., ceramic), such as a
stabilized zirconia, for example, YSZ, SSZ, etc.
[0016] Due to finite manufacturing tolerances, it is common to
insert compliant layers between the interconnect and cells to
support both the sealing and the electrical contact function, as
described in U.S. application Ser. Nos. 10/369,322 and 10/369,133
filed on Feb. 20, 2003, which are incorporated herein by reference
in their entirety.
[0017] A compliant seal material in the form of a paste, a tape, or
a gasket provides sealing of the interconnect to the cell. A
compliant conductive material, which can be applied as a paste, for
example, provides an electrical conduction path from the
interconnect to the cell.
[0018] The inventors have realized that these compliant layers can
also at least temporarily serve as structural elements in the
stack. If the compliant layer has the properties of an adhesive it
can provide mechanical bonding between the interconnect and the
cell.
[0019] Generally the compliant layers are prepared from the actual
functional material (e.g. metal powder, metal oxide powder,
glass-ceramic seal material, etc.) and a combination of solvents
and binders that render the active material suitable for
application. Application can happen by a variety of methods.
[0020] The compliant layer may be cast into a tape, which is then
with a use of an adhesive attached to the interconnect or the cell.
If an adhesive is supplied to both sides of the compliant tape, a
mechanical connection between interconnect and cell can be
achieved. This mechanical connection can provide structural
integrity to the stack.
[0021] The compliant layer can act as a contact glue. In this case,
the adhesive (which also fulfills the function of contact layer
and/or seal) is applied to both mating surfaces
(electrolyte/cell/frame and interconnect) (not necessarily the same
adhesive on both sides) and the bond is created when the two
adhesive layers contact the cell and the interconnect,
respectively. Thus, the contact adhesive acts as the compliant
layer when the stack and interconnect are adhered together.
[0022] Another way of applying the compliant layers is in a form of
paste which is either printed or dispensed onto either (part of)
the interconnect or the cell or both. If the binder and solvent
used in the compliant layer can serve as an adhesive, and the stack
is assembled while the adhesive is active, a structural bond
between interconnect and cell can be created.
[0023] In one preferred aspect of the first embodiment, the
adhesive used in the compliant layer is thermally activated. The
thermally activated adhesive can be either reversible (e.g. melt)
or irreversible (e.g. chemical cross link). In this case, the
compliant layer can be applied to the interconnect or the cell or
both sides before stack assembly. When the stack is assembled with
these layers in place, it is heated to activate the adhesive and a
strong mechanical bond is formed. For example, the adhesive can be
melted or cross linked to form the bond between the cell and
interconnect. The softening of the compliant layers during melting
can provide improved compliance.
[0024] In another aspect of the first embodiment, the compliant
layer is inserted as a solid (e.g. mesh, foam, felt, tape) and is
activated as an adhesive during assembly. There are number of ways
to activate a solid compliant layer to act as an adhesive. One
possible way is by using a thermoplastic behavior of the solid body
or part of the solid body, where the solid body is melted or
partially melted. This type of solid compliant layer thus acts as a
heat activatable adhesive.
[0025] One advantage of using the compliant layer as an adhesive is
that the structural integrity of the fuel cell stack is improved. A
stack prepared in this fashion can easily be handled and
transported without the need or at least less need for fixtures to
keep parts aligned.
[0026] Quite often the compliant layer is engineered such that the
binder and solvent, which in this embodiment may provide the
function of an adhesive, is burned out during the initial heat up
of the stack and only the functional material remains. Thus, the
adhesive binder and/or solvent are removed from the stack prior to
its operation to generate electricity and the compliant layer may
functions as a temporary adhesive during stack assembly. However,
the active materials, such as the compliant tape or felt, may
proceed to form a new mechanical bond between adjacent components,
such as a bond between an interconnect and cell.
[0027] There is a large number of materials that can be used as
binders and solvents in the compliant layer described above. One
example for a thermally activated organic material is
polycarbonate.
[0028] Either the seal or the electrical contact layer or both can
be engineered to provide the adhesive function. In other words, the
seal may be formed around a periphery of the electrodes in each
cell to keep the fuel or air gas flow on or away from a fuel or air
electrode of the cell. The seal may comprise an electrically
insulating material. The contact layer is an electrically
conductive layer which contacts the respective electrode and
interconnect to provide an electrical contact between the electrode
and interconnect.
II. Failure Tolerant Fuel Cell Stacks
[0029] The second embodiment of the present invention describes
various ways to avoid failure of the fuel cell stack in case of a
failure of a single fuel cell within the stack.
[0030] Fuel cells are usually operated in series, and depending on
the application, large numbers of cells can be connected. Should
one cell within a series of cells fail, then the internal
resistance in this failed cell frequently increases to large
values. In the most extreme case, all the voltage generated within
the stack can be lost within the resistance in one cell. Even if
total failure of the stack is not reached, significant losses of
performance can occur and the heat dissipated in the failed cell
can affect neighboring cells and cause the failure to
propagate.
[0031] The prior art approach to this problem is to identify and
eliminate the failure mechanism and thereby preserve the full
performance potential of a series of fuel cells. However, often the
cause of failure cannot be predicted or there is no remedy
available once cell failure occurs.
[0032] Some modern high-performance cars have a design philosophy
that can be applied to fuel cells. These cars often do not contain
spare tires, due to a lack of space. Instead, the tires are
designed such that they allow slow driving (also called "limp
home") even if a tire gets damaged.
[0033] In a fuel cell stack, a similar principle can be used. The
first aspect of the first embodiment of the invention addresses
failures that affect the anode (i.e., the fuel electrode) of a fuel
cell. In many fuel cells, the anode is the primary cause for
catastrophic failure. Some anodes require strong reducing
conditions in order to maintain their electrochemical, electrical,
and mechanical functions. In case of insufficient fuel supply or
excessive local currents (this applies primarily to oxygen
conducting electrolytes of solid oxide fuel cells), the anode can
be exposed to an environment that is not reducing enough, thus
resulting in possible irreversible damage. One example for this
type of failure is a cermet anode inside a solid oxide fuel cell.
In this case, the anode is composed of a mixture of metal and
ceramic, for example yttria stabilized zirconia (YSZ) and nickel.
If the nickel is exposed to too high oxygen partial pressure it
oxidizes and its conductivity is greatly reduced. During the
oxidation the nickel changes volume (it expands) and the expansion
can cause the anode structure to be destroyed (e.g. delamination of
the anode). This kind of failure can quickly reduce the available
conduction path for the ions traveling inside the electrolyte and
the electrons transported away from the electrode. In this case, a
single cell that was previously operating at a positive voltage
(e.g. 0.75V) can become so resistive that its voltage becomes
inversed and very large. This increase in resistive losses is only
limited by the available voltage of all cells in series in the
stack, in which case all power generated is lost in one single
cell. This power is largely dissipated via ohmic heating with can
lead to mechanical failure of the cell or adjacent
interconnects.
[0034] The inventors have realized that materials can be added to
the anode that can provide conductivity even when the actual anode
is damaged. These materials may be unsuitable to be used by
themselves due to performance and/or cost constraints, but a small
amount of this material within the electrode can maintain a finite
conductivity and thereby limit the losses that can occur within one
cell. Examples for these "emergency conductors" are chromium (or
its oxides), titanium oxides, or platinum. These examples
illustrate several mechanisms by which conductivity can be
maintained. Some of the mechanisms which can provide emergency
conductivity are: [0035] metallic conductor that does not oxidize;
[0036] conductive oxide, nitride and/or carbide; [0037] conductive
oxide that forms upon oxidation and reaction with the metal phase;
[0038] non-conductive oxide that becomes conductive (e.g. by
application of large enough potential); [0039] reduction and/or
dielectric break down; [0040] resistor parallel to cell.
[0041] These mechanisms may be used alone or in any suitable
combination with one or more of the other mechanisms. For the first
mechanism, a metallic conductor that does not oxidize, noble metals
including Pd, Pt, Ag and Au and their respective alloys may be
used. However, operating and processing temperatures may limit the
use of some Ag and Au alloys. In this case, the noble metal may be
added to the cermet, such as a nickel-stabilized zirconia
cermet.
[0042] The second mechanism relies on a conductive oxide to provide
the electrical conduction in the anode during reoxidation
conditions. Examples of conductive oxides include, but are not
limited to, Cr.sub.2O.sub.3; Ce, Gd, or Sm-doped Cr.sub.2O.sub.3;
Gd or Sm-doped CeO.sub.2; La--Cr perovskite phases including
La.sub.1-xSr.sub.xCr.sub.1-yMn.sub.yO.sub.3; Nb-doped TiO.sub.2; Ce
or Ti doped ZrO.sub.2; Mn--Cr--Co based spinels; and Gd--Ti--Mo
based pyrochlores. In addition, some conductive carbides or
nitrides such as TiN could function as the conductive phase. The
conductive oxide may be added to the cermet, such as a
nickel-stabilized zirconia cermet, or it may be used as the ceramic
phase of the cermet instead of the stabilized zirconia or it may be
used as the entire anode material instead of the cermet.
[0043] In the third mechanism, a conductive oxide forms upon
oxidation of the nickel metal phase. For example, small
introductions of La- or Ti-oxide to nickel metal will result in the
formation of the conductive perovskite phases LaNiO.sub.3 and
NiTiO.sub.3 upon local oxidation. Similarly, the nickel metal can
be doped such that upon oxidation a doped and more conductive
nickel oxide phase is formed. For example, M (Fe and/or Co) doping
of nickel metal will result in Ni.sub.1-xM.sub.xO.sub.2 upon
oxidation. In this mechanism, the anode cermet contains La oxide or
Ti oxide in addition to the nickel and stabilized zirconia and/or
it contains Fe and/or Co alloyed nickel or a mixture of Ni and Fe
and/or Co as the metal phase.
[0044] In the fourth and fifth mechanisms, a non-conductive oxide
or material becomes electrically conductive in a partially reducing
atmosphere and with a large enough electrochemical potential.
Examples of the oxides include ZrO.sub.2 and CeO.sub.2. Subsequent
doping of these binary oxides with a mixed valence ion, such as Ti
or Co, can significantly reduce the breakdown voltage and increase
the electrical conductivity. In this mechanism, the anode cermet
contains Ti and/or Co doped zirconia and/or ceria in addition to
the nickel metal phase.
[0045] In an alternative aspect of the second embodiment, a bypass
conductor is formed between adjacent interconnects in case of cell
failure using the latter mechanisms described above. Commonly used
solid oxide fuel cells (SOFC) often contain components needed to
incorporate one of these mechanisms. The electrolytes used in SOFCs
are usually ceramic oxygen ion conductors which become electronic
conductors if large electrical potentials are applied. YSZ, which
is one of the most popular electrolyte materials for SOFCs, reduces
at potentials near 2.2V. If a thin sheet of YSZ is mated between
two metallic surfaces, this assembly will have very poor total
conductivity unless potentials higher than the reduction potential
(near 2.2V) are applied at the metallic surfaces. Above the
reduction potential, the YSZ is partly or completely reduced to
metallic zirconium and becomes an electron conductor with much
lower resistance. If this assembly is parallel to a fuel cell (it
can also be integrated into the cell assembly, but still behave
electrically parallel) the current is allowed to bypass the cell
once the reduction potential is reached.
[0046] It is important to note that the reduction potential will
only be reached once the cell resistance is so large that a large
enough inverse (to normal operating polarity) potential has built
up.
[0047] The amount of conductivity available from this bypass
depends on a number of factors and can be engineered to meet the
needs of the cell. Some of the important parameters are: [0048]
available conduction area [0049] thickness of ceramic layer [0050]
choice of ceramic material [0051] choice and bonding method of
mating materials (these do not necessarily have to be metallic
conductive oxides) may also provide attractive properties.
[0052] FIG. 1 shows a sketch of one non-limiting embodiment of the
bypass conductor. A thin layer of YSZ 101 sprayed onto the cathode
side of the metallic interconnect 103A is the central part of the
bypass conductor 105. The spraying ensures that the cathode side of
this film is hermetically bonded to the metal and thereby no oxygen
ions can be created on the cathode side. If oxygen ions can be
created at the cathode side there is a potential (small) leakage
current under normal cell operation. However, other suitable layer
deposition methods which provide a hermetic bond can also be used
instead of spraying.
[0053] On the anode side of the layer 101, a thin layer 107 of
porous lanthanum strontium manganite (LSM) provides an escape path
(triple phase boundaries) for oxygen ions, which can help the
reduction process at high reverse potentials (removal of oxygen
from the YSZ).
[0054] As shown in FIG. 1, this bypass conductor 105 comprising
layers 101 and 107 can become a part of the fuel cell assembly, but
it could also be realized with a separate part which is
electrically connected to the cell. In other words, as shown in
FIG. 1, the anode and/or cathode side interconnect (such as the
anode side interconnect 103B shown in FIG. 1) contains a peripheral
extension 109 which extends toward the opposite interconnect 103A
laterally or peripherally of the cell boundaries (i.e., around one
or more edges of the electrolyte and the electrodes). The extension
may extend outside the seals 110 as well. The YSZ 101 and LSM 107
layers physically separate the interconnects from each other at the
periphery of the cell. Also shown in the FIG. 1 is the fuel cell
containing electrodes 113, 115 separated by an electrolyte 117. The
fuel cell is located between interconnects 103A and 103B. The
cathode electrode 113 electrically contacts interconnect 103A while
the anode electrode 115 of the cell electrically contacts
interconnect 103B. An electrolyte 117, such as a solid oxide
electrolyte of a SOFC, is located between the electrodes.
[0055] Alternatively, a separate device comprising two metal layers
separated by LSM and YSZ layers may be used to form a bypass
conductor. Each metal layer is electrically connected to a
respective anode and cathode side interconnect. Thus, the LSM and
YSZ layers may be located between separate metal layers inside or
outside the stack boundary rather than between peripheral portions
of the interconnects.
[0056] During normal operation of the stack, the anode side of the
cell is negative and the cathode is positive. During cell failure
the polarity reverses.
[0057] During normal operation, oxygen ions will try to migrate
through the YSZ bypass 101 from the top to the bottom of FIG. 1.
However they cannot escape at the hermetically sealed bottom
interface and thereby only a negligible current can be sustained.
As shown in FIG. 1, the bypass conductor 105 is located outside the
fuel cell in hot ambient air. The LSM layer 107 is porous so air
(oxygen) can travel through the porous LSM to the LSM/YSZ
boundary.
[0058] During cell failure, oxygen ions try to migrate through the
YSZ bypass 101 from the bottom to the top. A gradient in oxygen ion
concentration will be established inside the YSZ, which once it is
large enough, leads to reduction of the YSZ starting at the cathode
side metallic interface. This reduction will continue all the way
to the anode side interface in which case the metallic
interconnects, the reduced YSZ, and the mixed conduction LSM form a
bypass circuit. In this way, the current can flow between the
cathode side 103A and anode side 103B interconnect through the
bypass conductor 105 which comprises the reduced YSZ 101 (i.e., the
conductive Zr metal formed during the reduction) and the
electrically conductive LSM 107.
[0059] Other solid state oxide ionic conductors including Ce-
and/or Sc-doped ZrO.sub.2; and Sm-, Y-, and/or Gd-doped CeO.sub.2
could function as the bypass material 101 instead of or in addition
to YSZ. In addition, double doped fluorite systems including
Y.sub.xTi.sub.yZr.sub.1-x-yO.sub.2 could also be used.
[0060] Furthermore, other electrically conductive perovskites, such
as LSCo, LSCr, etc., and other electrically conductive ceramics may
be used instead of LSM 107.
III. Low Cost SOFC Interconnect
[0061] The third embodiment provides a low cost, multi-component
interconnect by using common sheet metal forming techniques.
[0062] The interconnect is comprised of three components: a flow
separator plate, a cathode contact media and an anode contact
media. The flow separator plate's purpose is to keep the fuel and
air flow streams from intermixing. The contact media preferably
comprise inserts which are inserted in or fitted to the flow
separator plate. The cathode insert's purpose is to expose or
provide an air flow to the fuel cell cathode (i.e., air) electrode
while maintaining electrical conductivity. Any suitable material
which can provide these functions at the fuel cell operating
temperature may be used. For example, corrugated steel or a high
temperature alloy foam may be used, as will be described in more
detail in the fifth embodiment. The anode insert's purpose is to
expose a fuel flow to the anode (i.e., fuel) electrode while
maintaining electrical conductivity. Any suitable material which
can provide these functions at the fuel cell operating temperature
may be used. For example, corrugated steel or a nickel foam may be
used, as will be described in more detail in the fourth
embodiment.
[0063] The interconnect may be used with any suitable fuel cells,
such as solid oxide fuel cells. The interconnect functions as a gas
separator plate between adjacent fuel cells while providing
electrical connection to and between the adjacent fuel cells. This
construction provides an inexpensive interconnect that fulfills all
the SOFC's requirements.
[0064] FIG. 2 illustrates the three components of the interconnect
201 prior to assembly. The flow separator plate (middle) part 203
is located between the anode 205 and cathode 207 inserts. The flow
separator plate 203 may contain side rails 209 or side clamps which
hold the inserts in place in the interconnect. While the
interconnect illustrated in FIG. 2 contains two riser openings 211
for an internally manifolded for fuel and externally manifolded for
air stack configuration, it should be understood that the
interconnect may contain four riser openings for an internally
manifolded for air and fuel configuration, or no riser openings for
an externally manifolded for air and fuel configuration. FIG. 3
illustrates the top view of the interconnect 201 after
assembly.
IV. Gas Flow Distribution Foam
[0065] As described with respect to the third embodiment above, an
interconnect may comprise an anode contact media which comprises an
electrically conductive foam material. The fourth embodiment
describes the details of this foam. It should be noted that while
the foam preferably comprises the anode insert in a flow separator
plate described in the third embodiment, the foam may be located in
a differently configured interconnect. For example, the
interconnect may lack the cathode insert or the interconnect may
comprise a monolithic structure rather than an insert-in-a-flow
separator plate design.
[0066] The foam material in the anode flow field provides both
electrical conductivity and allows gases (such as fuel gases) to
flow from the stack inlet to the stack outlet (i.e., from the fuel
inlet riser opening to the fuel outlet riser opening). Suitable
foam materials, such as nickel and nickel alloy foams are described
in U.S. application Ser. No. 10/369,133 filed on Feb. 20, 2003 and
incorporated herein by reference in its entirety. FIG. 4
illustrates a foam insert 205 located within the interconnect anode
cavity (i.e., attached to the anode contact side of the
interconnect's flow separator plate 203). The interconnect 201
contains the side rails 209 and fuel riser openings 211 described
above.
[0067] The present inventors noted that it is desirable to increase
a pressure drop in the anode flow field foam to optimize in plane
flow fuel distribution, since there are locations within the fuel
cell cavities where there are high gas utilizations. The present
inventors realized forming areas of low and high pressure will
direct the gas flow to all areas of the fuel cell. This provides an
ability to control the location of the gases within the
interconnect cavity.
[0068] Preferably, the foam contains surface features which direct
the gas flow in a desired location and/or direction over the anode
electrode of the fuel cell. The surface features may comprise
protrusions, such as ribs, and/or depressions, such as grooves to
direct the fuel flow.
[0069] The surface features may be formed by cutting and pressing
the foam into a die to impart the desired surface features to the
foam. Other foam patterning methods may also be used to form the
features.
[0070] FIG. 5 illustrates one non-limiting example of surface
features 213 which comprise bowed or curved ribs, such as
arc-shaped, semi-circular or semi-oval ribs. The ribs are concave
facing the fuel inlet riser opening (i.e., the fuel inlet) and are
convex facing the fuel exhaust riser opening (i.e., the fuel
outlet). The regions between the ribs comprise similarly shaped
grooves. Alternatively, the foam may contain grooves formed into
the surface of the foam rather than ribs extending from the surface
of the foam to form an equivalent structure.
[0071] FIG. 6 illustrates a CFD modeling of gas distribution within
an anode half cavity in which the foam contains the ribs shown in
FIG. 5. The modeling illustrates the velocity in inches per
second.
V. Foam Alloy for SOFC Cathode Contact and Flow Media
[0072] As described with respect to the third embodiment above, an
interconnect may comprise a cathode contact media which comprises
an electrically conductive foam material. The fifth embodiment
describes the details of this foam. It should be noted that while
the foam preferably comprises the cathode insert in a flow
separator plate described in the third embodiment, the foam may be
located in a differently configured interconnect. For example, the
interconnect may lack the anode insert or the interconnect may
comprise a monolithic structure rather than an insert-in-a-flow
separator plate design.
[0073] The interconnect contains a foam material in its cathode
flow field which provides both electrical conductivity and allows
air to flow from the stack inlet to the stack outlet.
[0074] Nickel foam may be utilized on the anode side of the
interconnect. However, since the cathode side of the interconnect
is maintained in an oxidizing ambient, a nickel foam on the cathode
side of the interconnect would oxidize very quickly.
[0075] Therefore, the cathode side foam comprises a material that
can withstand the oxidation rates within a high temperature, moist,
oxidizing environment without losing its electrical conductivity.
The foam preferably comprises an oxidation resistant nickel alloy
foam, such as a nickel alloy that contains nickel and one or more
of oxidation resistant alloying elements, such as Cr, W, Co and/or
Mo.
[0076] For example, the foam alloy may comprise a Haynes 230 alloy
foam which has the following composition.
TABLE-US-00001 Nominal Chemical Composition, Weight Percent Ni Cr W
Mo Fe Co Mn Si Al C La B 57.sup.a 22 14 2 3* 5* 0.5 0.4 0.3 0.10
0.02 0.015*
[0077] In general nickel alloys which contain 0 to 35, such as 10
to 30 weight percent Cr, 0 to 25, such as 10 to 20 weigh percent W,
0 to 5 weight percent Mo, 0 to 5, such as 1 to 4 weight percent Co
and at least 45, preferably greater than 50 weight percent Ni may
be used. Preferably, the alloy contains at least one non-zero
weight percentage of Cr, W or Mo. The alloy may also optionally
comprise 0 to 5 weight percent Fe and 0 to 1 weight percent of one
or more of Mn, Si, Al, C, La and B.
[0078] The foam is very flat, compliant and would allow gas flow
though it. It also is electrically conductive. Alternatively, a
felt or a mesh made of the nickel alloy may be used instead of the
foam.
VI. Ceramic Interconnect with Flow Channels and Conducting Vias
[0079] The sixth embodiment of the invention provides a ceramic
interconnect for use in a series connected, planar solid oxide fuel
cell stack. The one piece ceramic interconnect is multifunctional
and provides: a) manifolding and flow field for air, b) manifolding
and flow field for fuel, c) gas separation between two chambers,
and d) electrical interconnection of planar fuel cells.
[0080] The interconnect (IC) in a planar solid oxide fuel cell
serves multiple functions, namely: 1) manifolding of air and fuel,
2) gas separation, and 3) electrical connection. In addition, the
IC should possess adequate mechanical integrity, should be stable
at SOFC operating temperatures, should have a similar coefficient
of thermal expansion (CTE) of the cells, and should not react
chemically with either the electrodes or sealing materials.
Typically, ICs for planar SOFC stacks are metal-based and
fabricated by a number of standard forming processes. Metal-based
ICs have proven to be highly successful and have some inherent
advantages including high thermal conductivity, good mechanical
strength, and a potentially low cost fabrication route. However,
one of the main disadvantages of metal-based ICs is the formation
of a highly resistive oxide scale at operating temperature which
can significantly limit the life of a planar SOFC stack. An
alternative approach to the metallic IC is a ceramic-based IC that
has the same functionality, yet eliminates the growth of an oxide
scale with time.
[0081] A ceramic IC of the sixth embodiment incorporates channels
into the structure for flow fields and utilizes metal filled vias
for electronic conduction. An exemplary geometry of the
ceramic-based IC is shown in FIGS. 7 through 11. FIG. 7 is a top
view, FIGS. 8 and 9 are three dimensional views and FIGS. 10 and 11
are side cross sectional views along lines A-A and B-B in FIG. 9,
respectively. FIG. 7 is a schematic illustration of a top view of
ceramic IC with flow fields and conducting vias. FIG. 8 is a three
dimensional schematic illustration of a ceramic IC showing the top
two layers (1 and 2). FIG. 9 is a three dimensional schematic
illustration of flow fields and conducting vias of the ceramic
IC.
[0082] In the illustrated case, the IC is formed from five distinct
ceramic layers that are laminated together in the green state to
form the final geometry. The top two layers 1 and 2 comprise the
flow field for the cathode side of the cell, layer 3 is the solid
gas-separator layer, and layers 4 and 5 comprise the flow fields
for the anode side of the cell.
[0083] In this example, the flow fields are formed by punching
offset rectangular holes 11 and 12, 14 and 15 in two adjacent
layers (1 and 2; 4 and 5 respectively).
[0084] The offset hole-pattern results in an over-under flow path
with an a-square dictated by the width of the channels and the
thickness of the layers 1 and 2. The air and fuel flow directions
are illustrated in FIG. 10 by the dashed and the dashed-dotted
lines, respectively. As shown, the gases flow above each respective
interior layer 2, 4 and below each respective exterior layer 1,
5.
[0085] In this design, the geometry of the flow fields for the air
and fuel side of the interconnect can be designed independently for
optimal flow distribution. In other words, the hole size and/or
distribution may be different on the air and fuel side of the IC.
The electrical current is carried through the IC by an array of
metal or other electrically conductive material filled vias 6
located in between the flow channels. The vias 6 are punched in the
green tape and filled with metal by a screen printing process. The
5-layer green part is laminated together and sintered, resulting in
a fully dense ceramic part with metal via pathways.
[0086] In the schematic shown in FIGS. 7-9, the channels 11 in
layer 1 are 0.050'' wide and 0.2'' long, the offset channels 12 in
layer 2 are 0.050'' wide and 0.1'' long, the overlap is 0.050'' and
the thickness of each layer is 0.015''. The metal vias in this
example part are 0.010'' in diameter and are aligned through all
five layers. The large openings 21, 22 are the fuel riser
openings.
[0087] However, this is only one possible example and the channel
dimensions including width, length, thickness, overlap, spacing,
and general shape can be widely varied and are only limited by the
fabrication methods. Likewise, the vias are not limited to circular
shapes and can be varied in dimensions.
[0088] The ceramic-based IC can be fabricated from a number of
materials including fully stabilized zirconia, partially stabilized
zirconia, alumina, alumina-zirconia composites, such as alumina-YSZ
or alumina-SSZ composites, and MgO--SiO.sub.2 based materials
including forstertite. The layers 1-5 can be made of the same or
different ceramic materials. Suitable metals for the vias include
Pt, Pd, Au, Ag or their alloys, or electrically conductive ceramic
materials, such as LSM, LSCo or other conductive perovskite
materials. The ceramic materials can be made by roll compaction or
other ceramic fabrication techniques.
[0089] Various alternative configurations are also possible, as
described below.
[0090] Offset Vias: The via containing the metal does not have to
be co-axial in the z-direction, but can be offset, as described in
U.S. application Ser. No. 10/822,707 filed on Apr. 13, 2004 and
incorporated herein by reference in its entirety. For example, a
via can be co-axial through layers 1 to 3, offset by a given amount
in the x or y direction, and then be co-axial through layers 4 and
5. The two offset vias are connected electrically by a screen
printed pad that resides between layers 3 and 4. This configuration
allows for the use of differing metals in the vias for the air and
fuel sides of the IC. For example, Pt metal can fill the air side
of the offset via and Ni metal can fill the fuel side of the offset
via. The offsetting of the vias ensures that the vias are hermetic
if porosity develops in one of the metal fillers.
[0091] Conductive Oxide IC: The ceramic IC could be fabricated from
an electrically conductive oxide, such as LSM, LSC or other
conductive perovskite materials. Furthermore, as noted above, the
vias could be filled with an electrically conductive oxide instead
of a metal. Still further, the over-under gas flow configuration
described herein may also be provided in a metal rather than a
ceramic interconnect.
[0092] Catalysts: Catalysts can be added to the fuel side of the IC
for internal reforming of hydrocarbon based fuels. The catalyst
could be an insert in the channels, a coating on the channel
surfaces, or incorporated into the ceramic material itself. For
example, layer 5 of a 5-layer IC could be fabricated from a
NiO/zirconia composite that upon reduction would result in a
Ni-zirconia cermet.
[0093] Gas Manifolding: In the example shown in FIGS. 7-9, the fuel
is internally manifolded and the air is externally manifolded, in a
counter-flow geometry. However, any combination of
internal-internal, internal-external, or external-external
manifolding is possible. Likewise, flow geometries including
counter-flow, co-flow, and cross-flow are possible with the IC
design.
[0094] Number of Layers: While a five layer IC is illustrated in
FIGS. 7-9, the IC may contain more or less than 5 layers. For
example, layers 2 and/or 4 may be eliminated and substituted by
protrusions on one or both surfaces of layer 3. In this case, the
gases would flow over the protrusions in layer 3 and under the
respective layers 1 and 5. In another configuration, the IC may
comprise a single layer containing channels or tunnels extending
below the surfaces of the IC to allow the gas flow to go under
portions of the IC surfaces.
VII. SOFC Glass Fracturing Seal
[0095] The seventh embodiment of the invention provides a glass
seal containing a weak boundary layer which will controllably
fracture during thermal cycles. FIG. 12 illustrates an example of a
glass fracturing seal 310. For example, the fuel cell seal 310 may
be comprised of two glass 312, 314 layers with a layer of coarse
ceramic 316 located between the glass layers. This coarse layer
creates a weak boundary 318 between the layers of glass. In case of
a coefficient of thermal expansion (CTE) mismatch between the fuel
cell 301 and interconnect 303 materials, the weak layer will
fracture as the stack heats up and cools. This provides a
controlled fracture layer that doesn't harm the fuel cells during
thermal cycles. In other words, the weak boundary layer creates a
"crumple zone" which absorbs the CTE mismatch by fracturing, to
prevent or reduce the fracture of the fuel cells. FIG. 12 shows the
seal 310 located between a portion of the fuel cell 301, such as an
electrolyte or one of the electrodes, and a rib 306 of an
interconnect 303. The gas flow channels 304 are located between the
interconnect ribs 306.
[0096] The seal may be located between each fuel cell and an
adjacent interconnect in a fuel cell stack. The fuel cells are
preferably solid oxide fuel cells. However, the seal may be used
with other fuel cell types.
[0097] Any suitable glass materials may be used for the glass seal
layers 312, 314. The glass may be completely amorphous or it may
contain small ceramic or other particles (i.e., a glass-ceramic
material). Any suitable ceramic material may be used as the weak
layer 316. The ceramic weak layer preferably comprises large
particles which will provide a fracture plane in case of thermal
stresses.
[0098] Although the foregoing refers to particular preferred
embodiments, it will be understood that the present invention is
not so limited. It will occur to those of ordinary skill in the art
that various modifications may be made to the disclosed embodiments
and that such modifications are intended to be within the scope of
the present invention. All of the publications, patent applications
and patents cited herein are incorporated herein by reference in
their entirety.
* * * * *